Hybrid prediction model for a sales prospector

• US 8,775,230 B2
• Filed: 06/02/2009
• Issued: 07/08/2014
• Est. Priority Date: 11/03/2008
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Casas, Francisco V., Kim, Jooyoung John, Danko, Krisztian Z., Stengard, Peter J., Mozes, Ari, Campos, Marcos
• Primary Examiner(s): Pats, Justin M
• Assistant Examiner(s): Koneru, Sujay
• Application Number:
  US12/476,566
• Publication Number:
  US 20100114663A1
• Time in Patent Office: 1,862 Days
• Field of Search: 705 711- 742
• US Class Current: 705/7.31
• CPC Class Codes: G06F 16/214 Database migration support
• CPC Class Codes: G06F 16/2282 Tablespace storage structures; Management thereof
• CPC Class Codes: G06F 16/902 using more than one table in sequence, i.e. systems with three or more layers
• CPC Class Codes: G06Q 10/067 Enterprise or organisation modelling
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 30/0204 Market segmentation

Abstract

Systems and methods provide a system for generating a sales prospect recommendation that uses demographic data to make a sales prospect recommendation that includes a product recommendation with a probability that the sale will close, and may include an estimated time to close the sale and projected revenue. The system imports customer data including past purchasing data and demographic data for a plurality of customers. The system can then generate a predictive model by training the model using the past purchasing data and the demographic data. When queried for a sales prospect recommendation, the system responds to the query with at least one sales prospect recommended by the predictive model.

Claims

1. A non-transitory computer-readable medium having instructions stored thereon that, when executed by a processor, cause the processor to generate a sales prospect recommendation by:

uploading customer data including past purchasing data and demographic data for a plurality of customers, wherein the past purchasing data comprises a difference between a lead date and an order date for past transactions;

requesting a sales prospect recommendation;

receiving at least one sales prospect recommendation from a predictive model trained by the past purchasing data and the demographic data, wherein the at least one sales prospect recommendation comprises a prospective customer and product, and a probability that the prospective customer will purchase the product; and

displaying the sales prospect recommendation on a user interface wherein each prospective customer is represented by a bubble;

wherein the sales prospect recommendation further comprises an estimate of a time to close a sale of the product to the prospective customer, and an expected revenue from the sale;

wherein the estimate of the time to close the sale of the product to the prospective customer is based at least on differences between the lead dates and the order dates for past transactions for customers that are similar to the prospective customer;

wherein the user interface comprises a first axis of the probability that the prospective customer will purchase the product and a second axis of the estimate of the time to close the sale of the product to the prospective customer;

wherein the bubble is positioned on the first axis and the second axis, and a size of the bubble represents the estimated revenue from the sale.

2. The computer-readable medium of claim 1, wherein demographic data comprises at least one of:

a customer location, industry, headquarter country, public corporation status, importer status, exporter status, number of locations, number of employees, or annual revenue.

3. The computer-readable medium of claim 1, wherein the predictive model comprises an association rule model, a first cluster model, and a second cluster model.

4. The computer-readable medium of claim 3, wherein the first cluster model is built on the demographic data.

5. The computer-readable medium of claim 3, wherein the second cluster model is built on the past purchasing data.

6. The computer-readable medium of claim 3, wherein the association rule model is used to predict a product/customer pair and a probability that the sale will occur.

7. A computer-implemented method for generating a sales prospect recommendation, comprising:

receiving by a processor customer data including past purchasing data and demographic data for a plurality of customers, wherein the past purchasing data comprises a difference between a lead date and an order date for past transactions;

generating by the processor a predictive model trained by the past purchasing data and the demographic data;

receiving a query for a sales prospect recommendation;

responding to the query with at least one sales prospect recommendation based on the predictive model; and

causing the displaying of the sales prospect recommendation on a user interface;

wherein the at least one sales prospect recommendation comprises a prospective customer and product, and a probability that the prospective customer will purchase the product;

wherein the sales prospect recommendation further comprises an estimate of a time to close a sale of the product to the prospective customer, and an expected revenue from the sale;

wherein the estimate of the time to close the sale of the product to the prospective customer is based at least on differences between the lead dates and the order dates for past transactions for customers that are similar to the prospective customer;

wherein the user interface comprises a first axis of the probability that the prospective customer will purchase the product and a second axis of the estimate of the time to close the sale of the product to the prospective customer, and each prospective customer is represented by a bubble;

wherein the bubble is positioned on the first axis and the second axis, and a size of the bubble represents the estimated revenue from the sale.

8. The method of claim 7, wherein the predictive model includes an association rule model, a first cluster model, or a second cluster model.

9. The method of claim 8, wherein the first cluster model is built on the demographic data.

10. The method of claim 8, wherein the second cluster model is built on the past purchasing data.

11. The method of claim 8, wherein the association rule model comprises an Apriori algorithm.

12. A system for generating a sales prospect recommendation, comprising:

a processor;

computer readable media coupled to the processor storing instructions that implement a sales modeler when executed by the processor, and storing a customer data file;

the customer data file including demographic data and past purchasing data for a plurality of customers, wherein the past purchasing data comprises a difference between a lead date and an order date for past transactions;

the sales modeler trained using the customer data file that receives a prospective customer name as an input and outputs a product name and a probability that a prospective customer corresponding to the customer name will purchase a product corresponding to the product name; and

a user interface generated at least in part by the processor for querying the sales modeler;

wherein the sales modeler further outputs an estimate of a time to close a sale of the product to the prospective customer, and an expected revenue from the sale;

wherein the estimate of the time to close the sale of the product to the prospective customer is based at least on differences between the lead dates and the order dates for past transactions for customers that are similar to the prospective customer;

the user interface comprising a first axis of the probability that the prospective customer will purchase the product and a second axis of the estimate of the time to close the sale of the product to the prospective customer;

wherein at least one bubble for each prospective customer is positioned on the first axis and the second axis, and a size of the bubble represents the estimated revenue from the sale.

13. The system of claim 12, wherein the sales modeler includes an association rule model, a first cluster model, and a second cluster model.

14. The system of claim 13, wherein the first cluster model is built on the demographic data.

15. The system of claim 13, wherein the second cluster model is built on the past purchasing data.

16. The system of claim 13, wherein the association rule model comprises an Apriori algorithm.

17. A non-transitory computer-readable medium having instructions stored thereon that, when executed by a processor, cause the processor to generate a sales prospect recommendation by:

uploading customer data including past purchasing data and demographic data for a plurality of customers, wherein the past purchasing data comprises a difference between a lead date and an order date for past transactions;

requesting a sales prospect recommendation; and

receiving at least one sales prospect from a predictive model trained by the past purchasing data and the demographic data, wherein the at least one sales prospect comprises a prospective customer and product, wherein the predictive model includes an association rule model, a first cluster model built on the demographic data, and a second cluster model built on the past purchasing data, and wherein the first and second cluster models are used to predict a probability that the prospective customer will purchase the product, an estimated time to close the sale and an estimated revenue from the sale;

displaying the sales prospect recommendation on a user interface wherein each prospective customer is represented by a bubble;

wherein the estimate of the time to close the sale of the product to the prospective customer is based at least on differences between the lead dates and the order dates for past transactions for customers that are similar to the prospective customer;

wherein the user interface comprises a first axis of the probability that the prospective customer will purchase the product and a second axis of the estimate of the time to close the sale of the product to the prospective customer;

wherein the bubble is positioned on the first axis and the second axis, and a size of the bubble represents the estimated revenue from the sale.